Flow-enablement of the NASA SensorWeb using RESTful (and secure) workflows

Applied Remote Sensing to Disaster Management requires just-in-time delivery of custom data products to unsophisticated end-users such as fire fighters and first responders. These requests are wide-ranging, from wild fire hot spots detection, smoke, fire suppression and rehabilitation, flood coverage to erupting volcanoes, and require coordination of many assets such as satellites, UAV´s and ground sensors. They also require predictive models to complete the SensorWeb. Using Open Geospatial Consortium (OGC) standards, distributed web services from many organizations have been geo-enabled (or sensor-web enabled) to process the data and distribute it to end-users using Web 2.0 technologies such as Atom feeds and KML. The next challenge is to flow-enable these services to facilitate automated orchestration for on-demand requests coming from various communities in times of need. With our desire to provide the ability to create quick mash-ups for our end-users using a simple web browser, we have implemented a RESTful architecture and applied it to workflow management with the help of the Workflow Management Coalition (WfMC) and the OGC to support interoperability across our SensorWeb Community. These workflows must operate on behalf of a wide range of users and access services that may or may not be restricted. At a minimum, data consumers and providers need to communicate over http using some level of authentication that can be easily implemented in a RESTful manner. This paper also presents our Decision Support System used to manage our various Communities of Interests (COI) and give them transparent access to the SensorWeb assets and relevant custom data products when the needs arise. An Analytical Hierarchy Process (AHP) is used to automate the daily selection of requests to be scheduled for imaging and processing based on themes of interest.